A Delicate Balance

Clinical trial seeks to define optimal oxygen levels for premature babies

Oxygen is the most commonly used therapy in the neonatal intensive care unit: about 90% of infants will need oxygen at some point during their stay.

For decades, clinicians have debated how much oxygen should be given to preterm infants to avoid delivering too much in treating lung disease and other respiratory illnesses, which could cause long-term vision and neurological impairment.

“Oxygen is a drug, and like any drug, you need to know the risks and benefits of any intervention that you provide a patient, especially a preterm infant. Optimal oxygen saturation ensures that any risks of extra oxygen being given to the baby are outweighed by the benefits, which would be promoting growth, minimizing the impact of morbidity and, consequently, having a positive impact on the neurodevelopmental outcomes of these children,” says Dr. Elizabeth Asztalos, a clinician-scientist in the Women & Babies Research Program at Sunnybrook Research Institute (SRI). “We’re trying to balance those risks and benefits, so that the benefits outweigh the risks.”

Kelsie Cote in the neonatal intensive care unit (NICU) at Sunnybrook. About 90% of babies in the NICU will need oxygen therapy during their stay.

Photo: Doug Nicholson

Asztalos and colleagues conducted the Canadian Oxygen Trial, led by Dr. Barbara Schmidt of the University of Pennsylvania, and published the results in The Journal of the American Medical Association in 2013. They compared the effects of targeting oxygen saturations at a lower level of 85% to 89% compared with a higher level of 91% to 95% in extremely preterm infants. The aim was to determine the rate of death or disability at 18 months’ corrected age, the child’s age from the expected date of delivery, calculated by subtracting the number of weeks born before 40 weeks’ gestation from chronological age. The trial enrolled 1,201 infants aged 23 to 27 weeks’ gestation at 25 hospitals in Canada, the U.S., Argentina, Finland, Germany and Israel over four years. The babies were monitored until a postmenstrual age of 36 to 40 weeks (gestational plus chronological age).

Results found no significant effect of oxygen level on the rate of death or disability between the two saturation groups. “What surprised me was that we didn’t see a difference in cognition and motor skills, and we didn’t see the same impact on mortality as seen in the U.S. and other trials,” says Asztalos, who is also director of the Centre for Mother, Infant and Child Research at SRI and an associate professor at the University of Toronto. The other trials found the rate of death greater among the lower-target group versus the higher-target group.

Despite the controversy around the results of these trials, she says the study may still help guide clinicians in setting parameters for preterm infants.

Asztalos also notes that the Canadian study was able to meet its full sample size as planned and to continue with assessment to 18 to 21 months’ corrected age. Sunnybrook monitored 103 infants: 93 from Sunnybrook, and 10 from other Canadian sites, for the trial and the neurodevelopmental follow-up. "As a member of the steering committee, we can take pride in how we did the trial, and that we were very strong from a methodological perspective.”

— Eleni Kanavas

The Canadian Institutes of Health Research funded this research.